Uni-LVC: A Unified Method for Intra- and Inter-Mode Learned Video Compression

2026-03-05

AI Summary

This work addresses the limitations of existing learned video compression methods, which typically require separate training of intra- and inter-frame models and suffer significant performance degradation when reference frames are unreliable. The authors propose Uni-LVC, a unified framework that models inter-frame coding as intra-frame coding conditioned on temporal information from reference frames, thereby supporting both low-latency and random-access modes within a single model. The approach incorporates a cross-attention adaptation module to fuse temporal cues, a reliability-aware classifier to dynamically modulate the strength of temporal information, and a multi-stage training strategy for adaptive multimode learning. Experimental results demonstrate that Uni-LVC achieves superior rate-distortion performance over state-of-the-art methods in both intra- and inter-frame configurations while maintaining comparable computational efficiency.

Technology Category

Computer Vision: Multi-modal VisionMachine Learning: Multimodal LearningSearch and Optimization: Learning to Search

Application Category

Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphsSemantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMsUser Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization

Abstract

Recent advances in learned video compression (LVC) have led to significant performance gains, with codecs such as DCVC-RT surpassing the H.266/VVC low-delay mode in compression efficiency. However, existing LVCs still exhibit key limitations: they often require separate models for intra and inter coding modes, and their performance degrades when temporal references are unreliable. To address this, we introduce Uni-LVC, a unified LVC method that supports both intra and inter coding with low-delay and random-access in a single model. Building on a strong intra-codec, Uni-LVC formulates inter-coding as intra-coding conditioned on temporal information extracted from reference frames. We design an efficient cross-attention adaptation module that integrates temporal cues, enabling seamless support for both unidirectional (low-delay) and bidirectional (random-access) prediction modes. A reliability-aware classifier is proposed to selectively scale the temporal cues, making Uni-LVC behave closer to intra coding when references are unreliable. We further propose a multistage training strategy to facilitate adaptive learning across various coding modes. Extensive experiments demonstrate that Uni-LVC achieves superior rate-distortion performance in intra and inter configurations while maintaining comparable computational efficiency.

Problem

• learned video compression
• intra-mode coding
• inter-mode coding
• temporal reference reliability
• unified compression model

Innovation

• learned video compression
• unified coding framework
• cross-attention adaptation
• reliability-aware temporal modeling
• multistage training